Single Snapshot Multi-angle Target 2D Reconstruction based on RANSAC and RAM

Single snapshot multi-angle fusion radar target reconstruction refers to the real-time reconstruction of the target by using the observation data of a single snapshot and combining the radar observation data of multiple angles. Based on the geometric projection relationship of the target scattering center in the radar line-of-sight, this paper uses the random sample consensus (RANSAC) method to realize the detection of the validity of the scattering center correlation samples, associate the scattering centers based on multi-angle high resolution range profile (HRRP). Next, the multi-station data is applied to the association output of RANSAC for the target reconstruction process, which is very sensitive to the distance and angle measurement error of multiple stations. To deal with distance error, the reweighted atomic norm minimization (RAM) to realize the super-resolution. The angle measurements error is calibrated by the fused centroid as the distance alignment reference point. The bistatic geometry between every two stations further improves the reconstruction accuracy; hence, the number of stations can be reduced. The performance of the proposed methods is verified by the simulation results.